Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “accurate quantification”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions

Machine learning (ML) models, and Long Short-Term Memory (LSTM) networks in particular, have demonstrated remarkable performance in streamflow prediction and are increasingly being used by the hydrological research community. However, most of these applications do not include uncertainty quantification (UQ). ML models are data driven and can suffer from large extrapolation errors when applied to changing climate/environmental conditions. UQ is required to quantify the influence of data noises on model predictions and avoid overconfident projections in extrapolation. In this work, we integrate a novel UQ method, called PI3NN, with LSTM networks for streamflow prediction. PI3NN calculates Prediction Intervals by training 3 Neural Networks. It can precisely quantify the predictive uncertainty caused by the data noise and identify out-of-distribution (OOD) data in a non-stationary condition to avoid overconfident predictions. We apply the PI3NN-LSTM method in the snow-dominant East River Watershed in the western US and in the rain-driven Walker Branch Watershed in the southeastern US. Results indicate that for the prediction data which have similar features as the training data, PI3NN precisely quantifies the predictive uncertainty with the desired confidence level; and for the OOD data where the LSTM network fails to make accurate predictions, PI3NN produces a reasonably large uncertainty indicating that the results are not trustworthy and should avoid overconfidence. PI3NN is computationally efficient, robust in performance, and generalizable to various network structures and data with no distributional assumptions. It can be broadly applied in ML-based hydrological simulations for credible prediction.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Sensitivity of Optical Satellites to Estimate Windthrow Tree-Mortality in a Central Amazon Forest

Windthrow (i.e., trees broken and uprooted by wind) is a major natural disturbance in Amazon forests. Images from medium-resolution optical satellites combined with extensive field data have allowed researchers to assess patterns of windthrow tree-mortality and to monitor forest recovery over decades of succession in different regions. Although satellites with high spatial-resolution have become available in the last decade, they have not yet been employed for the quantification of windthrow tree-mortality. Here, we address how increasing the spatial resolution of satellites affects plot-to-landscape estimates of windthrow tree-mortality. We combined forest inventory data with Landsat 8 (30 m pixel), Sentinel 2 (10 m), and WorldView 2 (2 m) imagery over an old-growth forest in the Central Amazon that was disturbed by a single windthrow event in November 2015. Remote sensing estimates of windthrow tree-mortality were produced from Spectral Mixture Analysis and evaluated with forest inventory data (i.e., ground true) by using Generalized Linear Models. Field measured windthrow tree-mortality (3 transects and 30 subplots) crossing the entire disturbance gradient was 26.9 ± 11.1% (mean ± 95% CI). Although the three satellites produced reliable and statistically similar estimates (from 26.5% to 30.3%, p < 0.001), Landsat 8 had the most accurate results and efficiently captured field-observed variations in windthrow tree-mortality across the entire gradient of disturbance (Sentinel 2 and WorldView 2 produced the second and third best results, respectively). As expected, mean-associated uncertainties decreased systematically with increasing spatial resolution (i.e., from Landsat 8 to Sentinel 2 and WorldView 2). However, the overall quality of model fits showed the opposite pattern. We suggest that this reflects the influence of a relatively minor disturbance, such as defoliation and crown damage, and the fast growth of natural regeneration, which were not measured in the field nor can be captured by coarser resolution imagery. Our results validate the reliability of Landsat imagery for assessing plot-to-landscape patterns of windthrow tree-mortality in dense and heterogeneous tropical forests. Satellites with high spatial resolution can improve estimates of windthrow severity by allowing the quantification of crown damage and mortality of lower canopy and understory trees. However, this requires the validation of remote sensing metrics using field data at compatible scales.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Numerical Simulation of Infrasound Resonance in Underground Tunnel Structures

Remote observation of infrasound resonant signals emanating from underground tunnel structures could potentially allow the remote quantification of the geometry of the underground structures in which the signals were generated. However, the sensitivity of these observations to tunnel geometry and changes in that geometry are unknown. In this report we outline a numerical simulation study with the following three objectives: 1. Can we model infrasound resonance using numerical simulations? 2. What is the sensitivity of the modeled observations to changes in tunnel geometry or boundary conditions? And 3. Can we accurately model resonant observations of explosions occurring in Redmond Salt Mine in Central Utah. In this report we outline affirmative answers to the first two objectives, but we were unable to accurately model the Redmond Explosions due to numerical instability in the model of the complex structure of the mine. Recommendations for future work emphasize the need to acquire additional datasets and to explore a more data-based approach in which changes in data signatures are detected as a first step towards developing a method which inverts resonant infrasound signals for tunnel geometry.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Data-driven projection pursuit adaptation of polynomial chaos expansions for dependent high-dimensional parameters

Uncertainty quantification (UQ) and inference involving a large number of parameters are valuable tools for problems associated with heterogeneous and non-stationary behaviors. The difficulty with these problems is exacerbated when these parameters are statistically dependent requiring statistical characterization over joint measures. Probabilistic modeling methodologies stand as effective tools in the realms of UQ and inference. Among these, polynomial chaos expansions (PCE), when adapted to low-dimensional quantities of interest (QoI), provide effective yet accurate approximations for these QoI in terms of an adapted orthogonal basis. These adaptation techniques have been cast as projection pursuits in Gaussian Hilbert space in what has been referred to as a projection pursuit adaptation (PPA) by Xiaoshu Zeng and Roger Ghanem (2023). The PPA method efficiently identifies an optimal low-dimensional space for representing the QoI and simultaneously evaluates an optimal PCE within that space. The quality of this approximation clearly depends on the size of the training dataset, which is typically a function of the adapted reduced dimension. Here, the complexity of the problem is thus mediated by the complexity of the low-dimensional quantity of interest and not the complexity of the high-dimensional parameter space.

Data-driven↗
DOE OSTI2025

Bayesian inference of nuclear-matter density from proton scattering

Background: Proton elastic scattering at intermediate energy is widely employed as a tool for determining the matter radius of atomic nuclei. Here, the sensitivity of the approach relies on high-resolution measurements at small scattering angles and low-momentum transfer. Under these conditions, the Glauber multiple scattering theory accurately describes the proton-nucleus elastic cross section. Purpose: Investigate the sensitivity of the Glauber multiple scattering theory to uncertainties associated with input parameters such as the nuclear-matter density distribution and nucleon-nucleon data. Method: A joint Bayesian inference was performed using 12 angular distributions of elastic scattering at different energies on 58 Ni, 90 Zr, and 208 Pb targets. A Metropolis-Hastings algorithm was implemented to make an uncertainty quantification analysis for the input parameters used in the Glauber multiple scattering theory. Results: The experimental cross sections were fitted simultaneously using a joint Bayesian inference approach. Posterior probability density distributions of 42 input parameters were obtained from the analysis. A moderate correlation between the nuclear density parameters and the nucleon-nucleon cross sections was found. This correlation impacts the extraction of the nuclear-matter radius. Conclusions: The present analysis provided a consistent method for extracting the nuclear-matter density distribution of 58 Ni, 90 Zr, and 208 Pb from data across different incident energies. Due to the correlation of the nucleon-nucleon cross sections with the other input parameters, a constrained Bayesian inference using free nucleon-nucleon cross section data was performed. The nuclear-matter radii obtained from the analysis are in good agreement with multiple results reported in the literature.

190 ≤ A ≤ 219↗
DOE OSTI2022

Creep anisotropy modeling and uncertainty quantification of an additively manufactured Ni-based superalloy

The advantages offered by additive manufacturing over traditional processes has driven a great deal of industrial and academic interest in recent years. However, the process is relatively new and requires additional investigation to become sufficiently mature for wide scale industrial adoption. Electron beam melting powder bed fusion is one technology that has shown promise for fabricating high temperature resistant materials such as nickel based superalloys. The resulting microstructures typically exhibit a strong fiber texture in the build direction giving rise to anisotropic time-dependent deformation behavior. In order to accelerate the qualification of these materials for industrial adoption accurate numerical models are needed for simulating their behavior. In this work a crystal plasticity model including non-Schmid effects is presented for capturing creep anisotropy observed in additively manufactured IN738LC. The model is calibrated via a probabilistic framework where model parameters are treated as random variables. An iterative sequential design strategy is utilized to efficiently identify the probability density of the unknown model parameters. As a case study the model is utilized to investigate the behavior of randomly oriented equiaxed grain clusters sometimes observed embedded in the additively manufactured columnar structure. A synthetic realization is simulated and uncertainty is propagated through to the full-field response. Results indicate that these features are the source of significant creep relaxation and strain accumulation which partially explains observed grain boundary decohesion at these locations.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Multilevel Monte Carlo methods for the Grad-Shafranov free boundary problem

The equilibrium configuration of a plasma in an axially symmetric reactor is described mathematically by a free boundary problem associated with the celebrated Grad-Shafranov equation. The presence of uncertainty in the model parameters introduces the need to quantify the variability in the predictions. This is often done by computing a large number of model solutions on a computational grid for an ensemble of parameter values and then obtaining estimates for the statistical properties of solutions. In this study, we explore the savings that can be obtained using multilevel Monte Carlo methods, which reduce costs by performing the bulk of the computations on a sequence of spatial grids that are coarser than the one that would typically be used for a simple Monte Carlo simulation. We examine this approach using both a set of uniformly refined grids and a set of adaptively refined grids guided by a discrete error estimator. Numerical experiments show that multilevel methods dramatically reduce the cost of simulation, with cost reductions typically on the order of 60 or more and possibly as large as 200. Furthermore, adaptive griding results in more accurate computation of geometric quantities such as x-points associated with the model.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Boosting efficiency and reducing graph reliance: Basis adaptation integration in Bayesian multi-fidelity networks

The computational cost of high-fidelity numerical models makes outer-loop analysis, which requires repeated interrogation of the model such as uncertainty quantification, computationally demanding. Multi-fidelity methods, which construct a surrogate model using data from an ensemble of models of varying cost and accuracy, can substantially reduce the cost of outer-loop analysis. However, these methods can be difficult to apply when the model ensemble does not admit a clear hierarchy a priori and the correlations between models are low. Consequently, in this paper, we present a multi-fidelity method that leverages dimension reduction to enhance the correlation between models, thereby reducing the amount of data needed to train a surrogate from an unordered ensemble of models. Our method utilizes basis adaptation to build low-dimensional polynomial chaos expansions of each model and employs Multi-fidelity Networks to encode the relationships among models. We show that the resulting method exhibit two notable advantages over its counterpart: (1) enhanced accuracy (both reduced bias and variance); and (2) reduced dependency on the graph structure encoding relationships among models. We demonstrate the approach on an analytical test problem and a challenging finite element model for a spent nuclear fuel. Our method produces a surrogate model that is significantly more accurate than either a single-fidelity surrogate or a multi-fidelity surrogate constructed without basis adaptation.

42 ENGINEERING↗
DOE OSTI2020

Conditional Karhunen-Loève expansion for uncertainty quantification and active learning in partial differential equation models

We use a conditional Karhunen-Lo` eve (KL) model to quantify and reduce uncertainty in a stochastic partial differential equation (SPDE) problem with partially-known space-dependent coefficient, Y (x). We assume that a small number of Y (x) measurements are available and model Y (x) with a KL expansion. We achieve reduction in uncertainty by conditioning the KL expansion coefficients on measurements. We consider two approaches for conditioning the KL expansion: In Approach 1, we condition the KL model first and then truncate it. In Approach 2, we first truncate the KL expansion and then condition it. We employ the conditional KL expansion together with Monte Carlo and sparse grid collocation methods to compute the moments of the solution of the SPDE problem. Uncertainty of the problem is further reduced by adaptively selecting additional observation locations using two active learning methods. Method 1 minimizes the variance of the PDE coefficient, while Method 2 minimizes the variance of the solution of the PDE. We demonstrate that conditioning leads to dimension reduction of the KL representation of Y (x). For a linear diffusion SPDE with uncertain log-normal coefficient, we show that Approach 1 provides a more accurate approximation of the conditional log-normal coefficient and solution of the SPDE than Approach 2 for the same number of random dimensions in a conditional KL expansion. Furthermore, Approach 2 provides a good estimate for the number of terms of the truncated KL expansion of the conditional field of Approach 1. Finally, we demonstrate that active learning based on Method 2 is more efficient for uncertainty reduction in the SPDE’s states (i.e., it leads to a larger reduction of the variance) than active learning using Method 2.

Conditioned Karhunen-Lo` eve expanion, machine lea↗
DOE OSTI2022

Building surrogate models of nuclear density functional theory with Gaussian processes and autoencoders

From the lightest Hydrogen isotopes up to the recently synthesized Oganesson (Z = 118), it is estimated that as many as about 8,000 atomic nuclei could exist in nature. Most of these nuclei are too short-lived to be occurring on Earth, but they play an essential role in astrophysical events such as supernova explosions or neutron star mergers that are presumed to be at the origin of most heavy elements in the Universe. Understanding the structure, reactions, and decays of nuclei across the entire chart of nuclides is an enormous challenge because of the experimental difficulties in measuring properties of interest in such fleeting objects and the theoretical and computational issues of simulating strongly-interacting quantum many-body systems. Nuclear density functional theory (DFT) is a fully microscopic theoretical framework which has the potential of providing such a quantitatively accurate description of nuclear properties for every nucleus in the chart of nuclides. Thanks to high-performance computing facilities, it has already been successfully applied to predict nuclear masses, global patterns of radioactive decay like β or γ decay, and several aspects of the nuclear fission process such as, e.g., spontaneous fission half-lives. Yet, predictive simulations of nuclear spectroscopy—the low-lying excited states and transitions between them—or of nuclear fission, or the quantification of theoretical uncertainties and their propagation to basic or applied nuclear science applications, would require several orders of magnitude more calculations than currently possible. However, most of this computational effort would be spent into generating a suitable basis of DFT wavefunctions. Such a task could potentially be considerably accelerated by borrowing tools from the field of machine learning and artificial intelligence. In this paper, we review different approaches to applying supervised and unsupervised learning techniques to nuclear DFT.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Deep Learning and Likelihood Approaches for Viral Phylogeography Converge on the Same Answers Whether the Inference Model Is Right or Wrong

Abstract Analysis of phylogenetic trees has become an essential tool in epidemiology. Likelihood-based methods fit models to phylogenies to draw inferences about the phylodynamics and history of viral transmission. However, these methods are often computationally expensive, which limits the complexity and realism of phylodynamic models and makes them ill-suited for informing policy decisions in real-time during rapidly developing outbreaks. Likelihood-free methods using deep learning are pushing the boundaries of inference beyond these constraints. In this paper, we extend, compare, and contrast a recently developed deep learning method for likelihood-free inference from trees. We trained multiple deep neural networks using phylogenies from simulated outbreaks that spread among 5 locations and found they achieve close to the same levels of accuracy as Bayesian inference under the true simulation model. We compared robustness to model misspecification of a trained neural network to that of a Bayesian method. We found that both models had comparable performance, converging on similar biases. We also implemented a method of uncertainty quantification called conformalized quantile regression that we demonstrate has similar patterns of sensitivity to model misspecification as Bayesian highest posterior density (HPD) and greatly overlap with HPDs, but have lower precision (more conservative). Finally, we trained and tested a neural network against phylogeographic data from a recent study of the SARS-Cov-2 pandemic in Europe and obtained similar estimates of region-specific epidemiological parameters and the location of the common ancestor in Europe. Along with being as accurate and robust as likelihood-based methods, our trained neural networks are on average over 3 orders of magnitude faster after training. Our results support the notion that neural networks can be trained with simulated data to accurately mimic the good and bad statistical properties of the likelihood functions of generative phylogenetic models.

Evolutionary Biology↗
DOE OSTI2021

Flow-driven spectral chaos (FSC) method for simulating long-time dynamics of arbitrary-order non-linear stochastic dynamical systems

Uncertainty quantification techniques such as the time-dependent generalized polynomial chaos (TD-gPC) use an adaptive orthogonal basis to better represent the stochastic part of the solution space (aka random function space) in time. However, because the random function space is constructed using tensor products, TD-gPC-based methods are known to suffer from the curse of dimensionality. Here, we introduce a new numerical method called the flow-driven spectral chaos (FSC) which overcomes this curse of dimensionality at the random-function-space level. The proposed method is not only computationally more efficient than existing TD-gPC-based methods but is also far more accurate. The FSC method uses the concept of enriched stochastic flow maps to track the evolution of a finite-dimensional random function space efficiently in time. To transfer the probability information from one random function space to another, two approaches are developed and studied herein. In the first approach, the probability information is transferred in the mean-square sense, whereas in the second approach the transfer is done exactly using a new theorem that was developed for this purpose. The FSC method can quantify uncertainties with high fidelity, especially for the long-time response of stochastic dynamical systems governed by ODEs of arbitrary order. Six representative numerical examples, including a nonlinear problem (the Van-der-Pol oscillator), are presented to demonstrate the performance of the FSC method and corroborate the claims of its superior numerical properties. Finally, a parametric, high-dimensional stochastic problem is used to demonstrate that when the FSC method is used in conjunction with Monte Carlo integration, the curse of dimensionality can be overcome altogether.

(nonlinear) stochastic dynamical systems↗
DOE OSTI2023

A Unified Metric for Fast Frequency Response in Low-Inertia Power Systems

Future power systems with more inverter-based resources (IBRs), will be vulnerable to frequency decline contingencies. Fast frequency response (FFR) provided by IBRs is a good candidate to arrest frequency excursions. Diverse types of FFR have been proposed, and some have been deployed in our power systems. Without a unified quantification of FFR, it is hard for the grid operators to compare and fully leverage the FFR capabilities of IBRs. This work introduces a potential unified metric that quantifies two key characteristics of FFR and describes its application to three prevailing FFR types. We then use metric-to-frequency mapping to validate the accuracy of the metric in predicting the impact of a given FFR on the trajectory of a frequency event. The results show that the proposed metric is simple yet accurately captures the ability of diverse forms of FFR to improve system frequency dynamics.

effective inertia↗
DOE OSTI2021

A Bayesian interpretation of inconsistency measures in cosmology

Abstract Measures of inconsistency and tension between datasets have become an essential part of cosmological analyses. It is important to accurately evaluate the significance of such tensions when present. We propose here a Bayesian interpretation of inconsistency measures that can extract information about physical inconsistencies in the presence of data scatter. This new framework is based on the conditional probability distribution of the level of physical inconsistency given the obtained value of the measure. We use the index of inconsistency as a case study to illustrate the new interpretation framework, but this can be generalized to other metrics. Importantly, there are two aspects in the quantification of inconsistency that behave differently as the number of model parameters increases. The first is the probability for the level of physical inconsistency to reach a threshold which drops with the increase of the number of parameters under consideration. The second is the actual level of physical inconsistency which remains rather insensitive to such an increase in parameters. The difference between these two aspects is often overlooked, which leads to a long-standing ambiguity: when a given inconsistency is found between two constraints, its “significance” seems to be lower when considered in a higher-dimensional parameter space. This ambiguity is resolved by the Bayesian interpretation we introduce in this work because the conditional probability distribution includes all the statistical information of the level of physical inconsistency. Finally, we apply the Bayesian interpretation to examine the (in)consistency between Planck versus the Cepheid-based local measurement, the Dark Energy Survey (DES), the Atacama Cosmology Telescope (ACT) and WMAP. We confirm and revisit the degrees of previous physical inconsistencies and show the stability of the new interpretation with respect to the number of cosmological parameters compared to the commonly used n-σ interpretation when applied to cosmological tensions in multi-parameter spaces.

Astronomy & Astrophysics↗
DOE OSTI2022

Automatic information extraction from childhood cancer pathology reports

The International Classification of Childhood Cancer (ICCC) facilitates the effective classification of a heterogeneous group of cancers in the important pediatric population. However, there has been no development of machine learning models for the ICCC classification. We developed deep learning-based information extraction models from cancer pathology reports based on the ICD-O-3 coding standard. In this article, we describe extending the models to perform ICCC classification. We developed 2 models, ICD-O-3 classification and ICCC recoding (Model 1) and direct ICCC classification (Model 2), and 4 scenarios subject to the training sample size. We evaluated these models with a corpus consisting of 29206 reports with age at diagnosis between 0 and 19 from 6 state cancer registries. Our findings suggest that the direct ICCC classification (Model 2) is substantially better than reusing the ICD-O-3 classification model (Model 1). Applying the uncertainty quantification mechanism to assess the confidence of the algorithm in assigning a code demonstrated that the model achieved a micro-F1 score of 0.987 while abstaining (not sufficiently confident to assign a code) on only 14.8% of ambiguous pathology reports. Our experimental results suggest that the machine learning-based automatic information extraction from childhood cancer pathology reports in the ICCC is a reliable means of supplementing human annotators at state cancer registries by reading and abstracting the majority of the childhood cancer pathology reports accurately and reliably.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Creating an Advanced Sensor Network to calculate real-time, mass-weighted flue gas composition and air heater leakage of a coal-fired utility boiler under dynamic operating conditions

Utilization of renewable energy sources to minimize the environmental impact of energy production has changed the way utility boilers operate, requiring frequent load cycling between full load and partial loads as low as 30%. Dynamic operation of coal-fired utility boilers significantly reduces boiler efficiency when compared to steady state at full load. Data-driven plant optimization has shown success with coal-fired utility boilers under dynamic operating conditions. The purpose of this work was to create an Advanced Sensor Network (ASN) to provide more extensive real-time data to inform dynamic plant optimization of Net Unit Heat Rate (NUHR). The ASN consists of gas sampling grids in the convective pass of the boiler and downstream of the air heater. These sampling grids allow for quantification of spatial variation of flue gas within the boiler and calculation of mass-weighted composition of flue gas through the combination of composition, velocity, and temperature measurements. The comparison of O 2 between the inlet and outlet of the air heater is used to calculate air leakage in real time. Flue gas composition and air heater leakage are both important factors in boiler efficiency and NUHR. Further, the results of this work support the value of mass-weighted averages for determining flue gas composition accurately. The measurements from the ASN show increased composition stratification during dynamic operation, with an average standard deviation 38% higher than observed during steady-state operation. Air heater leakage was also observed to increase from 2.8% to 5.1% following a load change. Prior to the installation of the ASN, these data would not have been available for dynamic control. These real-time data will be leveraged to calculate and optimize for NUHR during dynamic operation in future work.

42 ENGINEERING↗
DOE OSTI2024

Data Generation for Machine Learning Interatomic Potentials and Beyond

The field of data-driven chemistry is undergoing an evolution, driven by innovations in machine learning models for predicting molecular properties and behavior. Recent strides in ML-based interatomic potentials have paved the way for accurate modeling of diverse chemical and structural properties at the atomic level. The key determinant defining MLIP reliability remains the quality of the training data. A paramount challenge lies in constructing training sets that capture specific domains in the vast chemical and structural space. This Review navigates the intricate landscape of essential components and integrity of training data that ensure the extensibility and transferability of the resulting models. We delve into the details of active learning, discussing its various facets and implementations. We outline different types of uncertainty quantification applied to atomistic data acquisition and the correlations between estimated uncertainty and true error. The role of atomistic data samplers in generating diverse and informative structures is highlighted. Furthermore, we discuss data acquisition via modified and surrogate potential energy surfaces as an innovative approach to diversify training data. The Review also provides a list of publicly available data sets that cover essential domains of chemical space.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗